Loading…
13 models tracked from OpenAI.
OpenAI's flagship frontier model for the most complex coding and professional reasoning work, with a ~1M-token context window and configurable reasoning effort.
The smallest and fastest current GPT-5.5 model, built for high-volume, latency-sensitive workloads with a 400K context window.
Cheaper, faster sibling of GPT-5.5 that keeps reasoning and tool use while trading a little quality for dramatically lower latency and cost.
The highest-capability GPT-5.5 variant, using more reasoning compute for the most difficult problems where cost and latency are secondary concerns.
Lower-latency, lower-cost mini model in the GPT-5.4 line, tuned for coding, computer use, and subagents.
An earlier GPT-5 generation flagship, superseded by GPT-5.5; still available for workloads pinned to its behavior.
The cost-efficient mini variant of the legacy GPT-5.2 generation, kept available for low-cost pinned workloads.
A late-2025 GPT-5 generation model, now legacy, retained for reproducibility and pinned production deployments.
The first GPT-5 point release, now legacy; an incremental update over GPT-5 with improved reasoning and tool use.
The original GPT-5 launch model, now legacy; the foundation of the GPT-5 family with reasoning and long-context support.
The maximum-effort variant of the original GPT-5, now legacy; trades latency for the highest accuracy of the first GPT-5 generation.
Smallest, lowest-cost GPT-5.4 variant for latency- and cost-sensitive workloads, with a 400K context window.
Prior-generation GPT-5 reasoning model offering a ~1M-token context window at lower cost than GPT-5.5.
Latest tracked developments from OpenAI.